or Connect
AVS › AVS Forum › Audio › Audio theory, Setup and Chat › Simplified REW Setup and Use (USB Mic & HDMI Connection) Including Measurement Techniques and How To Interpret Graphs
New Posts  All Forums:Forum Nav:

Simplified REW Setup and Use (USB Mic & HDMI Connection) Including Measurement Techniques and How... - Page 50

post #1471 of 9584
Quote:
Originally Posted by JChin View Post

Hi Pres2play, after making the changes try shutting down or restart lappy. Your laptop may require the reset for the changes to take effect.

I tried that but it didn't work. It's a good idea, though. I recall it being mentioned earlier in the thread. I'll try to find that post.

Gear mentioned in this thread:

post #1472 of 9584
Quote:
Originally Posted by JChin View Post

Quote:
Originally Posted by AustinJerry View Post

Glad you received the mic.  The only calibration files you need are on the USB stick.  Make sure you use the one for a 90-degree mic orientation.

Great, so point the mic up toward the ceiling when I take measurement, correct?

 

Yes, that would be correct for general room response measurements.  There are some situations in which you would want to take a nearfield measurement of a speaker by pointing the mic directly at the driver (e.g. when gain-matching subs), but these are not the types of measurements we are discussing in this thread.

post #1473 of 9584
Quote:
Originally Posted by AustinJerry View Post

Yes, unfortunately it sounds like you have a bad mic.  The only other thing you might try is plugging it into a different computer, perhaps a friend's, or one at work?  This would rule out the laptop as being the issue.

Jerry, in the sound control panel, can you click on the Mic icon and see if there's a mute setting? I read that Windows can set the mic to mute when there's a change/install.
post #1474 of 9584
Quote:
Originally Posted by Pres2play View Post

Quote:
Originally Posted by AustinJerry View Post

Yes, unfortunately it sounds like you have a bad mic.  The only other thing you might try is plugging it into a different computer, perhaps a friend's, or one at work?  This would rule out the laptop as being the issue.

Jerry, in the sound control panel, can you click on the Mic icon and see if there's a mute setting? I read that Windows can set the mic to mute when there's a change/install.

 

Yes, there is a mute button.  I hope this is your issue!  wink.gif

 

post #1475 of 9584
Thanks, Jerry. I'll get back to you guys later.
post #1476 of 9584
Quote:
Originally Posted by Selden Ball View Post

Quote:
Originally Posted by jkasanic View Post

my current surrounds are just Speakercraft AIM Wide One's and AIM Wide 8's with both sets being in-ceiling.  Any thoughts on why Audyssey may have set my surround xovers so low (don't tell me it's in the FAQ already)! eek.gif

Don't forget that Audyssey's microphone is intended to be used with grazing-incidence audio (i.e. speakers which are close to 90 degrees off axis). Unfortunately, that means that in-ceiling speakers (which are more nearly on-axis to the microphone) probably can't be calibrated as accurately as floor-standing speakers.

In other words, more than the crossover frequency might be having problems. frown.gif

This is, of course, a good reason for using REW: to find out just how bad things really are in your listening environment.

 

Thanks...I saw your post in the Audyssey thread about directional tweeters and thought of this as well.  One question I had is that when I was running new speaker cable for my remote AV rack, I didn't really notice that I had different mids from rear surrounds.  I think I put the AIM Wide One's as the mids and the AIM8 One's as the rear surrounds.  Probably a question for another thread but I was just wondering if anyone had any thoughts on whether the AIM8 One's with the directional tweeters would be better in one location vs. the other?  The AIM Wide One's are marketed as offering more dispersion than the traditional AIM8 series.  Anyone have any thoughts?

post #1477 of 9584
Quote:
Originally Posted by AustinJerry View Post

Yes, unfortunately it sounds like you have a bad mic.  The only other thing you might try is plugging it into a different computer, perhaps a friend's, or one at work?  This would rule out the laptop as being the issue.

Yeah, looks like it's the mic. Doesn't work on my desk top, either. I connected a different mic via the Mic Input and the Sound Recorder plays just fine. I'll contact Herb and get a replacement. The mic probably got shaken up a bit too hard on its was over.

I'm actually relieved the problem isn't my laptop.
post #1478 of 9584
Quote:
Originally Posted by Pres2play View Post

Quote:
Originally Posted by kbarnes701 View Post

It\s odd that the mic is defective though. I can believe this with an out-of-the-box product, but Herb has calibrated this mic and it must have worked when he did that surely?  HST I can't think of anything else to suggest. If Windows and REW can both 'see it', then it ought to, well, just work.

Agreed. I'm hoping it's just some weird setting on the laptop. I'll know definitely when I plug into my PC tonight.

I have found I have to follow a strict sequence of events when I use REW and my mic.

 

1. Connect all cables to laptop and AVR.

2. Power on laptop.

3. Power on AVR

4. Open REW.

 

If I change that sequence, sometimes either the HDMI or the USB mic seem to not be found reliably. Probably just my own setup, but that sequence has worked for me every time. 

 

I also found that when I used a different USB port one time, Windows had to be re-adjusted for the levels in the sound setup dialogs. Using the same USB port ensures that the settings are 'sticky' (although I now take a few seconds to check each one before continuing, per Jerry's advice).

 

Good luck.

 

EDIT: just read your later posts, so ignore this. I will leave it here in case it's useful to anyone else.

post #1479 of 9584

Just for interest, here are some graphs I made yesterday when double-checking my Sub Distance Tweak adjustment, using REW (previously determined using OM). I found that the previously set distances were still measuring the best using REW - but I also experimented with using a XO to my dual Submersives of 110Hz (as recommended to me by Mark Seaton) instead of the 100Hz I used previously. The subs, for those familiar with Submersives are EQ'd by Audyssey Pro using Pgm1 and then switched to Pgm2 afterwards to apply a roughly 3dB boost from about 40Hz down - my usual mode of running them. The sub trim levels are left as Audyssey suggested them - ie the subs are not run 'hot' at all. Audyssey, of course, is ON and Dynamic EQ/DynVol are OFF.

 

 

This is Centre channel + Subs, using 110Hz XO (all graphs are 1/6th smoothing):

 

 

 

This is the same as the above (Centre channel) but with a XO of 100Hz. I find it incredible that changing the XO by such a small amount has made the difference shown and am wondering if it is an anomalous result. Next time I get the measuring gear out I will double check this, but this is the result from yesterday using a XO of 100Hz:

 

 

 

Just for clarity, this is the above two graphs combined into one:

 

 

Here is the RF + LF + Subs, XO of 110Hz:

 

 

 

And here is a combined average showing RF + LF + C + Subs, all with 110 Hz XO as before:

 

 

 

Comments welcome. Anyone got a comment on that +9dB rise starting at 65Hz?  

post #1480 of 9584

Now that I have demonstrated to myself how amazingly good REW really is, I feel I ought to make a contribution to John M for all his hard work and efforts. Is there a feeling for what is a good amount to donate?  What have you other guys done?

post #1481 of 9584
Quote:
Originally Posted by kbarnes701 View Post

Just for interest, here are some graphs I made yesterday when double-checking my Sub Distance Tweak adjustment, using REW (previously determined using OM). I found that the previously set distances were still measuring the best using REW - but I also experimented with using a XO to my dual Submersives of 110Hz (as recommended to me by Mark Seaton) instead of the 100Hz I used previously...

Comments welcome. Anyone got a comment on that +9dB rise starting at 65Hz?  

I think the rise you get at low frequencies is because you pass the all-channel signal as input, then disconnected all speakers except of one and the sub, so sub channel plays the sum of all other channels at low frequencies, while higher frequencies are still getting the power of only one channel. Could it be the case?
post #1482 of 9584

@ Keith,

 

Above 70Hz, the graph looks perfectly normal and quite good.  The curve below 70Hz looks somewhat different from posts you have made previously.  Any idea why?  Not that it is particularly bad, just different.

 

How do you get left+right+center to play all at the same time?  When you have a full 8-channels available in ASIO, can you select as many as you want to play at the same time? 

post #1483 of 9584
Quote:
Originally Posted by kbarnes701 View Post

Now that I have demonstrated to myself how amazingly good REW really is, I feel I ought to make a contribution to John M for all his hard work and efforts. Is there a feeling for what is a good amount to donate?  What have you other guys done?

 

I think it is a personal decision, and I'm sure John would be pleased with any contribution.  It's kind of you make the effort.

post #1484 of 9584
Quote:
Originally Posted by IgorZep View Post

Quote:
Originally Posted by kbarnes701 View Post

Just for interest, here are some graphs I made yesterday when double-checking my Sub Distance Tweak adjustment, using REW (previously determined using OM). I found that the previously set distances were still measuring the best using REW - but I also experimented with using a XO to my dual Submersives of 110Hz (as recommended to me by Mark Seaton) instead of the 100Hz I used previously...

Comments welcome. Anyone got a comment on that +9dB rise starting at 65Hz?  

I think the rise you get at low frequencies is because you pass the all-channel signal as input, then disconnected all speakers except of one and the sub, so sub channel plays the sum of all other channels at low frequencies, while higher frequencies are still getting the power of only one channel. Could it be the case?

 

Interesting hypothesis, Igor.  Recall that the REW signal is a mono signal, and is being routed only to the selected channels.  So I don't think there is actually any other signals that would be causing a cumulative effect.  I don't notice a similar phenomenon when I am measuring.

post #1485 of 9584
Quote:
Originally Posted by AustinJerry View Post

Interesting hypothesis, Igor.  Recall that the REW signal is a mono signal, and is being routed only to the selected channels.  So I don't think there is actually any other signals that would be causing a cumulative effect.  I don't notice a similar phenomenon when I am measuring.

I've seen such raise when I did the following... As my HDMI in notebook is not capable to pass multichannel for measuring surrounds I selected All Channel Mono on the receiver, disconnected other speakers and did the measurements. If I measured the surround with sub combined - the raise was there, exactly because of what I have explained.
post #1486 of 9584
Quote:
Originally Posted by kbarnes701 View Post

And here is a combined average showing RF + LF + C + Subs, all with 110 Hz XO as before:

 

 

 

Comments welcome. Anyone got a comment on that +9dB rise starting at 65Hz?  

Keith, very interesting phenomenon you have below 65Hz.  I would definitely measure again to rule out an anomaly.  WRT the graph above, is that an "average" of 3 different measurements (e.g. L+Sub, R+Sub, C+Sub) or per Jerry's question, did you manage to measure all together using ASIO?

post #1487 of 9584

One more comment for Keith:  when using the Camera icon to capture the graphs for posting, please select the "legend" option so we can see what smoothing has been applied.  Thanks!

post #1488 of 9584
Quote:
Originally Posted by IgorZep View Post

Quote:
Originally Posted by kbarnes701 View Post

Just for interest, here are some graphs I made yesterday when double-checking my Sub Distance Tweak adjustment, using REW (previously determined using OM). I found that the previously set distances were still measuring the best using REW - but I also experimented with using a XO to my dual Submersives of 110Hz (as recommended to me by Mark Seaton) instead of the 100Hz I used previously...

Comments welcome. Anyone got a comment on that +9dB rise starting at 65Hz?  

I think the rise you get at low frequencies is because you pass the all-channel signal as input, then disconnected all speakers except of one and the sub, so sub channel plays the sum of all other channels at low frequencies, while higher frequencies are still getting the power of only one channel. Could it be the case?

 

I hear what you are saying there Igor. And yes, it could be the case. Thinking out loud... the crossover is set at for LCR at 100Hz so everything from those three channels below 100Hz is being redirected to the sub. I guess the question is: when I am only playing the centre channel (for example) via REW, is the signal from L&R still present, despite not being heard?  If it is, then you are correct. I don't know of any way around that, easily.  With OmniMic, I could just send a signal, via CD, to the L channel, and then a separate one to the R channel and a separate one to the L channel - so there would be no all-channel signal to act the way you suggest. That could explain the difference between the graphs I get at the low end on my OM measurements, compared with those I see when I use REW.

 

Anyone else got any thoughts on that?  If Igor is right, then the low end measurements for L, R or C channel signals will always be wrong. 

post #1489 of 9584
Quote:
Originally Posted by AustinJerry View Post

@ Keith,

 

Above 70Hz, the graph looks perfectly normal and quite good.  The curve below 70Hz looks somewhat different from posts you have made previously.  Any idea why?  Not that it is particularly bad, just different.

 

How do you get left+right+center to play all at the same time?  When you have a full 8-channels available in ASIO, can you select as many as you want to play at the same time? 

 

See my reply to the interesting point Igor makes - you are right - the below 70Hz is different (and not quite as good) as the one I see in OM measurements.

 

The L+C+R is just an average. I don't even know if it is of any use or if you can use the Average feature that way. Anyone?

post #1490 of 9584
Quote:
Originally Posted by AustinJerry View Post

Quote:
Originally Posted by kbarnes701 View Post

Now that I have demonstrated to myself how amazingly good REW really is, I feel I ought to make a contribution to John M for all his hard work and efforts. Is there a feeling for what is a good amount to donate?  What have you other guys done?

 

I think it is a personal decision, and I'm sure John would be pleased with any contribution.  It's kind of you make the effort.

 

OK. In that case, for the sake of full disclosure, I am going to send him $50. Does that seem fair to you guys here? 

post #1491 of 9584
Quote:
Originally Posted by AustinJerry View Post

@ Keith,

 

Above 70Hz, the graph looks perfectly normal and quite good.  The curve below 70Hz looks somewhat different from posts you have made previously.  Any idea why?  Not that it is particularly bad, just different.

 

How do you get left+right+center to play all at the same time?  When you have a full 8-channels available in ASIO, can you select as many as you want to play at the same time? 

Sorry, Jerry, I missed that part of your question. The answer is no, unfortunately. Any two channels at a time.

post #1492 of 9584
Quote:
Originally Posted by IgorZep View Post

Quote:
Originally Posted by AustinJerry View Post

Interesting hypothesis, Igor.  Recall that the REW signal is a mono signal, and is being routed only to the selected channels.  So I don't think there is actually any other signals that would be causing a cumulative effect.  I don't notice a similar phenomenon when I am measuring.

I've seen such raise when I did the following... As my HDMI in notebook is not capable to pass multichannel for measuring surrounds I selected All Channel Mono on the receiver, disconnected other speakers and did the measurements. If I measured the surround with sub combined - the raise was there, exactly because of what I have explained.

 

That is slightly different though isn't it?  In your case the signal is still being output by the AVR but you cannot hear it because the speakers are disconnected. In my case, I am just selecting one channel plus the subs in REW - in that case I'd have expected that no other signal would be present from the AVR other than the one selected (plus the subs). BICBW.

post #1493 of 9584
Quote:
Originally Posted by jkasanic View Post

Quote:
Originally Posted by kbarnes701 View Post

And here is a combined average showing RF + LF + C + Subs, all with 110 Hz XO as before:

 

 

 

Comments welcome. Anyone got a comment on that +9dB rise starting at 65Hz?  

Keith, very interesting phenomenon you have below 65Hz.  I would definitely measure again to rule out an anomaly.  WRT the graph above, is that an "average" of 3 different measurements (e.g. L+Sub, R+Sub, C+Sub) or per Jerry's question, did you manage to measure all together using ASIO?

 

That 'hump' is always present since using REW - it is on every graph I have made. Yes, it is an average. There is, AFAIK, no way to test three channels simultaneously. I am not even sure if the 'average' (used in this way) is of any use. I included it to spark off the discussion really.

 

I can see the value of averaging several different mic placement measurements of 1 channel - but not sure if the way I did it has any value. All comments gratefully received ;)

post #1494 of 9584
Quote:
Originally Posted by AustinJerry View Post

One more comment for Keith:  when using the Camera icon to capture the graphs for posting, please select the "legend" option so we can see what smoothing has been applied.  Thanks!

 

Yes I did consider that - but when you have 30 measurements open the legend part is bigger than the actual graph. I did state the smoothing applied though. (It was 1/6th for that last lot, as per 'Jerry's Rules').

post #1495 of 9584
Quote:
Originally Posted by kbarnes701 View Post

Quote:
Originally Posted by AustinJerry View Post

One more comment for Keith:  when using the Camera icon to capture the graphs for posting, please select the "legend" option so we can see what smoothing has been applied.  Thanks!

 

Yes I did consider that - but when you have 30 measurements open the legend part is bigger than the actual graph. I did state the smoothing applied though. (It was 1/6th for that last lot, as per 'Jerry's Rules').

 

I can understand that.  I tend to open up a fresh set of measurements for each REW session, and then save the measurements by date.  For each date, I document what changed, i.e. why I am running a fresh set of measurements.  This way, I rarely have more than 8-10 measurements per session.

post #1496 of 9584
Quote:
Originally Posted by kbarnes701 View Post

That is slightly different though isn't it?  In your case the signal is still being output by the AVR but you cannot hear it because the speakers are disconnected. In my case, I am just selecting one channel plus the subs in REW - in that case I'd have expected that no other signal would be present from the AVR other than the one selected (plus the subs). BICBW.

You should select only one channel for measuring combined with sub response. If you select the center + sub, then you send +10dB signal to the sub plus receiver sends redirected bass from the center channel. So, you get more than 10dB raise in the sub range. The puzzle solved?
post #1497 of 9584
Quote:
Originally Posted by AustinJerry View Post

Quote:
Originally Posted by kbarnes701 View Post

Quote:
Originally Posted by AustinJerry View Post

One more comment for Keith:  when using the Camera icon to capture the graphs for posting, please select the "legend" option so we can see what smoothing has been applied.  Thanks!

 

Yes I did consider that - but when you have 30 measurements open the legend part is bigger than the actual graph. I did state the smoothing applied though. (It was 1/6th for that last lot, as per 'Jerry's Rules').

 

I can understand that.  I tend to open up a fresh set of measurements for each REW session, and then save the measurements by date.  For each date, I document what changed, i.e. why I am running a fresh set of measurements.  This way, I rarely have more than 8-10 measurements per session.

 

Yes, I normally do it the way you do too. But in this case, tweaking the sub distances a little at a time, I soon got a lot of measurements!  I guess I could have discarded all the useless ones and saved a new file with just the relevant ones.

post #1498 of 9584
Quote:
Originally Posted by kbarnes701 View Post

 

 

That 'hump' is always present since using REW - it is on every graph I have made. Yes, it is an average. There is, AFAIK, no way to test three channels simultaneously. I am not even sure if the 'average' (used in this way) is of any use. I included it to spark off the discussion really.

 

I can see the value of averaging several different mic placement measurements of 1 channel - but not sure if the way I did it has any value. All comments gratefully received ;)

 

Well, time to leverage REW to get at the root cause.  I would focus on the following set of measurements only from 15-300Hz, no smoothing.  Also recommend Prog1 on the subs, if that is what you were running when you calibrated.

 

- Sub1 only, Audyssey off and on

- Sub2 only, Audyssey off and on

- Both subs, Audyssey off and on

- Both subs + Left + Right, Audyssey off and on

- Both subs + Center, Audyssey off and on

 

Rather than post a bunch of graphs, have you set up an account on a file-sharing site like DropBox where you can post the MDAT file?  I would like to take a stab at analyzing the data, rather than nickel-and-dime you to death with different graph requests.

 

And as far as your averaging question, I think averaging is best used for multiple mic locations of the same speaker(s).  When you average different speaker resounses with one mic position, you are simply masking the different responses, which is what you don't want to do.  IMHO, of course.

post #1499 of 9584
Quote:
Originally Posted by IgorZep View Post

Quote:
Originally Posted by kbarnes701 View Post

That is slightly different though isn't it?  In your case the signal is still being output by the AVR but you cannot hear it because the speakers are disconnected. In my case, I am just selecting one channel plus the subs in REW - in that case I'd have expected that no other signal would be present from the AVR other than the one selected (plus the subs). BICBW.

You should select only one channel for measuring combined with sub response. If you select the center + sub, then you send +10dB signal to the sub plus receiver sends redirected bass from the center channel. So, you get more than 10dB raise in the sub range. The puzzle solved?

 

Sorry Igor, I am not following you. To measure the impact of altering the sub distances in an effort to optimise the splice around the XO, I have to have the Centre channel and the Sub running at the same time. This is how the Sub Distance Tweak has always been documented (originally by Mark Seaton of Seaton Sound).  Unless the C channel and the sub are being measured together, how would I be able to evaluate the combined effect of the speakers around the splice?  Or am I misunderstanding you?

 

I really want to understand this, so please bear with me. How would you do it?

 

This is exactly what I am doing:

 

I select the C channel in REW and also select the sub in REW. The test signal then plays through the C channel and the sub at the same time. The only other way I can see to do this is to select just the C channel (and no other) in REW and measure, and then select just the sub and measure. But that won't tell me how the two speakers interact when both are playing at the same time, which is what I need to know.

 

Thank you for your patience.

post #1500 of 9584
Quote:
Originally Posted by AustinJerry View Post

Quote:
Originally Posted by kbarnes701 View Post

 

 

That 'hump' is always present since using REW - it is on every graph I have made. Yes, it is an average. There is, AFAIK, no way to test three channels simultaneously. I am not even sure if the 'average' (used in this way) is of any use. I included it to spark off the discussion really.

 

I can see the value of averaging several different mic placement measurements of 1 channel - but not sure if the way I did it has any value. All comments gratefully received ;)

 

Well, time to leverage REW to get at the root cause.  I would focus on the following set of measurements only from 15-300Hz, no smoothing.  Also recommend Prog1 on the subs, if that is what you were running when you calibrated.

 

 

OK - that is easy enough. 15-300 is what I was looking at here for the evaluation of the splice, no smoothing. After evaluation of the various distance settings, it became obvious that the ones I was already using (10.6 and 9.6 ft) were still the best (no reason to believe otherwise as no speaker/sub changes have been made - I just wanted to see that REW gave me the same result as M, which it did). Consequently I just posted the full spectrum to show the difference (alleged) that changing the XO made.

 

It doesn't really matter if I use Pgm1 or Pgm 2 because the difference between them is always totally consistent and very simple.  Here it is, graphed, no smoothing 15-300Hz. Pgm2 simply raises the level by a predictable amount. In the graph, the brown line is, obviously, Pgm 2. As you can see, nothing above about 50Hz is changed.

 

 

 

 

 

This raises another, general, question. Other than for comparison purposes between state A and state B, (eg Audyssey ON, Audyssey OFF) I can't really see the point in measuring the system in a way in which I will never actually use it. Because I always use Pgm2, I measure with Pgm2 engaged, same as I measure with Audyssey engaged. Is this not right?

 

Quote:

 Sub1 only, Audyssey off and on

- Sub2 only, Audyssey off and on

- Both subs, Audyssey off and on

- Both subs + Left + Right, Audyssey off and on

- Both subs + Center, Audyssey off and on

 

I already have those measurements - they are the ones recommended in your Guide are they not? But again, other than to get a feel for what Audyssey is actually doing, what's the point of measuring with Audyssey off in order to try to determine the best sub delays for splice optimisation?

 

 

Quote:

Rather than post a bunch of graphs, have you set up an account on a file-sharing site like DropBox where you can post the MDAT file?  I would like to take a stab at analyzing the data, rather than nickel-and-dime you to death with different graph requests.

 

 

Sure. I have a Dropbox account. I will upload the mdat file to it for you to look at. Will post the url for you later.

 

Quote:
And as far as your averaging question, I think averaging is best used for multiple mic locations of the same speaker(s).  When you average different speaker resounses with one mic position, you are simply masking the different responses, which is what you don't want to do.  IMHO, of course.

 

 
OK - that is what I thought in reality. Thanks.
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Audio theory, Setup and Chat

Gear mentioned in this thread:

AVS › AVS Forum › Audio › Audio theory, Setup and Chat › Simplified REW Setup and Use (USB Mic & HDMI Connection) Including Measurement Techniques and How To Interpret Graphs